A question-and-answer application with an “ask-to-answer” feature is described. The ask-to-answer feature enables any user to solicit an answer to a question from one or more users. Upon soliciting one or more users for an answer to a particular question, a message may be directed to the one or more users. The message may include a copy of the text of the question and may provide a mechanism enabling the one or more users to pass on answering the question. Subsequent to the solicitation, the question page may include a notification with information about the solicitation, including information identifying the group of users who have been asked to answer the question and the number of times the one or more users has been asked to provide an answer.
|
1. A computer-implemented method comprising:
receiving user input associated with a requesting user, the user input including a solicitation for an answer to a question and one or more factors for generating a ranked list of recommended users to provide the answer;
training a first machine-learning model to generate a predicted quality score of an answer that a user will provide;
generating, by the first machine-learning model, predicted quality scores for a list of recommended users based on the one or more factors, the predicted quality scores usable to generate a ranked list of recommended users;
receiving a selected list of recommended users, wherein the selected list of recommended users is selected from the ranked list of recommended users based on the predicted quality scores of the list of recommended users;
predicting, by a second machine-learning model, a likelihood that a recommended user of the selected list of recommended users will provide an answer to the question if solicited by a notification, wherein the likelihood is used to define a subset of the selected list of recommended users;
facilitating a communication, via a notification, to the subset of the selected list of recommended users, the communication including the question; and
presenting a webpage that includes the question, and an indication that the subset of the selected list of recommended users have been solicited to answer the question.
20. A non-transitory computer-readable medium having stored thereon instructions that, when executed by one or more processors, cause the one or more processors to perform operations including:
receiving user input associated with a requesting user, the user input including a solicitation for an answer to a question and one or more factors for generating a ranked list of recommended users to provide the answer;
training a first machine-learning model to generate a predicted quality score of an answer that a user will provide;
generating, by the first machine-learning model, predicted quality scores for a list of recommended users based on the one or more factors, the predicted quality scores usable to generate a ranked list of recommended users;
receiving a selected list of recommended users, wherein the selected list of recommended users is selected from the ranked list of recommended users based on the predicted quality scores of the list of recommended users;
predicting, by a second machine-learning model, a likelihood that a recommended user of the selected list of recommended users will provide an answer to the question if solicited by a notification, wherein the likelihood is used to define a subset of the selected list of recommended users;
facilitating a communication, via a notification, to the subset of the selected list of recommended users, the communication including the question; and
presenting a webpage that includes the question, and an indication that the subset of the selected list of recommended users have been solicited to answer the question.
15. A system comprising:
one or more processors; and
a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including:
receiving user input associated with a requesting user, the user input including a solicitation for an answer to a question and one or more factors for generating a ranked list of recommended users to provide the answer;
training a first machine-learning model to generate a predicted quality score of an answer that a user will provide;
generating, by the first machine-learning model, predicted quality scores for a list of recommended users based on the one or more factors, the predicted quality scores usable to generate a ranked list of recommended users;
receiving a selected list of recommended users, wherein the selected list of recommended users is selected from the ranked list of recommended users based on the predicted quality scores of the list of recommended users;
predicting, by a second machine-learning model, a likelihood that a recommended user of the selected list of recommended users will provide an answer to the question if solicited by a notification, wherein the likelihood is used to define a subset of the selected list of recommended users;
facilitating a communication, via a notification, to the subset of the selected list of recommended users, the communication including the question; and
presenting a webpage that includes the question, and an indication that the subset of the selected list of recommended users have been solicited to answer the question.
2. The computer-implemented method of
3. The computer-implemented method of
4. The computer-implemented method of
5. The computer-implemented method of
presenting to the requesting user information identifying one or more groups of users to whom the requesting user can direct a solicitation for an answer to the question.
6. The computer-implemented method of
7. The computer-implemented method of
8. The computer-implemented method of
9. The computer-implemented method of
10. The computer-implemented method of
11. The computer-implemented method of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question.
12. The computer-implemented method of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
13. The computer-implemented method of
upon detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question, converting the solicitation for an answer to the question to a vote by the requesting user for the answer to the question provided by the one or more recommended users of the subset of the selected list of recommended users.
14. The computer-implemented method of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating a recommended user from the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the recommended user from the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
16. The system of
17. The system of
18. The system of
19. The system of
presenting to the requesting user information identifying one or more groups of users to whom the requesting user can direct a solicitation for an answer to the question.
21. The system of
22. The system of
25. The system of
26. The system of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question.
27. The system of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
28. The system of
upon detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question, converting the solicitation for an answer to the question to a vote by the requesting user for the answer to the question provided by the one or more recommended users of the subset of the selected list of recommended users.
29. The system of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating a recommended user from the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the recommended user from the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
30. The non-transitory computer-readable medium of
31. The non-transitory computer-readable medium of
32. The non-transitory computer-readable medium of
33. The non-transitory computer-readable medium of
presenting to the requesting user information identifying one or more groups of users to whom the requesting user can direct a solicitation for an answer to the question.
34. The non-transitory computer-readable medium of
35. The non-transitory computer-readable medium of
36. The non-transitory computer-readable medium of
37. The non-transitory computer-readable medium of
38. The non-transitory computer-readable medium of
39. The non-transitory computer-readable medium of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has passed on responding to the solicitation for an answer to the question.
40. The non-transitory computer-readable medium of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
41. The non-transitory computer-readable medium of
upon detecting an event indicating that one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question, converting the solicitation for an answer to the question to a vote by the requesting user for the answer to the question provided by the one or more recommended users of the subset of the selected list of recommended users.
42. The non-transitory computer-readable medium of
subsequent to the communication of the question to the subset of the selected list of recommended users, detecting an event indicating a recommended user from the subset of the selected list of recommended users has provided an answer to the question; and
communicating a message to the requesting user, the message indicating that the recommended user from the one or more recommended users of the subset of the selected list of recommended users has provided an answer to the question.
|
This application claims benefit of U.S. Provisional Patent Application No. 62/668,140, filed May 7, 2018, the contents of which are hereby incorporated by reference in their entirety.
The present disclosure generally relates to data processing techniques. More specifically, the present disclosure relates to methods and systems for enhancing the quality of user-generated content for presentation via a question-and-answer service by enabling users of the service to directly solicit an answer to a question from another user and/or from a group of users.
Many online or web-based applications and services rely partially, and in some cases almost entirely, on user-generated content—that is, content generated by the end users of the particular application or service. The success of such applications and services is heavily dependent upon the ability to attract and retain users who are willing to participate by generating content that is to be shared with other users. One particular type of application or service for which this holds true is an on-line question-and-answer service. In general, a question-and-answer service is a service that allows its end-users to post questions, post answers to questions, or simply browse the questions and answers that others have posted. From the perspective of its users, the usefulness of a question-and-answer service or application depends significantly on the extent to which meaningful answers are provided to the questions being posted.
In some embodiments, a computer-implemented method is provided. The method comprises identifying a message associated with a requesting user. When the message is identified, a group of recommended users are selected based on a credential and past response behavior. The method further comprises receiving user input associated with a requesting user, the user input identifying one or more recommended users to whom a message is to be communicated, the message including a solicitation for an answer to a question. The method further comprises communicating the message. The method further comprises presenting a web page including the question. The web page includes a notification indicating that the one or more recommended users have been solicited for an answer to the question and a number of solicitations for an answer to the question that the one or more recommended users has received.
In some embodiments, a system is provided. The system comprises one or more processors. The system further comprises a non-transitory computer-readable medium including instructions that, when executed by the one or more processors, cause the one or more processors to perform operations including the steps of the above method.
In some embodiments, a non-transitory computer-readable medium is provided. The computer-readable medium has stored thereon instructions that, when executed by one or more processors, cause the one or more processors to perform operations including the steps of the above method.
Some embodiments of the invention are illustrated by way of example, and not limitation, in the figures of the accompanying drawings, in which:
Methods and systems for enabling a user of a question-and-answer service to directly solicit an answer to a question from another user are described. In the following description, for purposes of explanation, numerous specific details are set forth in order to provide a thorough understanding of the various aspects of different embodiments of the present invention. It will be evident, however, to one skilled in the art, that the present invention may be practiced without these specific details.
A question-and-answer application or service provides an online forum where users can post questions, post answers to questions, or simply search for and review questions and answers that others have posted. To aid in the overall organization and presentation of information, a question can be associated with one or more topics representing, for example, a subject matter to which the question relates. Take for example the question, “What is the population of San Francisco?” This question may be assigned to the topic, “San Francisco.” As users post questions for others to answer, users can designate each question as being related to one or more existing topics, or users can create and designate new topics. As such, a question-and-answer application is an information repository of user-generated content that is organized and presented in a relatively simple and intuitive manner. An individual web page showing all questions related to a particular topic may be referred to as a topic page, whereas a web page showing an individual question, and any associated answers to the question, is generally referred to herein as a question page.
To encourage users to share and exchange information, and to increase overall participation in a question-and-answer application, a question-and-answer application may enable users to share information with one another. For example, the user interface of a question-and-answer application may include a user interface element (e.g., a button or link) that, when selected, prompts the viewing user to provide contact information (e.g., a name, an email address, a telephone number) of a user with whom the information is to be shared. Any number of communication mechanisms might be used to share information in this manner, including, but limited to: email, instant messaging, SMS or text messaging, some proprietary messaging system, as well as any other similar messaging systems. In some instances, the message communicated to the user will contain a copy of the actual content. In other instances, the message may simply include a selectable link (e.g., a Uniform Resource Identifier (URI) or Uniform Resource Locator (URL)) to the relevant content. In this manner, a requesting user may share with a second user and/or a group of users an unanswered question, a question for which there are one or more existing answers, a particular answer to a question, or some other related item of information.
In one particular scenario, a requesting user may share a question with a second user whom the requesting user believes to be particularly well-suited to provide an answer. Stated differently, a user may share a question with another user with the expectation that the receiving user provide an answer to the question. Alternatively or additionally, a user may share a question with a group of users, with the expectation that one or more users from the group will provide an answer. However, while sharing information in this manner is certainly useful, it may or may not actually encourage or persuade a user to take a particular action, such as to provide an answer to a question. For instance, if a user receives a private and direct message from another user, such that the message includes a question, or a link to a question, the recipient may choose to simply ignore the message and not provide an answer to the question.
Various aspects of the inventive methods and systems described herein aim to increase the likelihood that a user and/or a group of users, when requested to answer a particular question, will be encouraged to actually provide an answer. For example, consistent with some embodiments of the invention, when a user solicits an answer to a question from another user or a group of users, the message communicated to the other user or the group of users includes a specific call to action, requesting that the recipient of the message provide an answer to the question. In some instances, the message may be formatted to include a copy of the actual text of the question, a link to the question page for the question, as well as a user interface element, such as a text input box, in which the recipient can enter or input his or her answer to the question. In addition, consistent with some embodiments, when a requesting user solicits an answer to a question from a second user or a group of users, the solicitation is made public.
For example, subsequent to the solicitation, the question page for the question can include a notification displayed next to or near the question indicating that a particular user has been solicited for an answer to the question. The notification may also include, for example, a name and picture of the user who has been solicited to answer the question. Selecting the name and/or picture of the user who has been asked to answer the question may cause some portion of the user's profile or biographical information to be displayed. If several users have solicited a particular user for an answer to the same question, the notification may indicate the number of times that the particular user has been solicited to answer the question. In some embodiments, the notification may also identify the individual user or users who have solicited the recipient for an answer. This information may be displayed in connection with the notification, or as a result of detecting a certain event, such as a mouse-over event.
As another example, the question page for the question can include a notification indicating that a particular group of users have been solicited to answer the question. The group can be, for example, an interest group that provides a space within the question-and-answer service to discuss a particular interest. In this example, for an interest group that has been solicited to answer the question, the question page can display the name of the interest group and/or a picture associated with the interest group. As another example, the group can be users with similar backgrounds, where the users' backgrounds may be derived, for example, from questions the users have answered, topics and/or questions the users follow, other users that the users follow, and/or the users' profile information, among other things. In this example, the question page can display text indicating that the question has been directed to users having a particular background. For example, the question page can display the string “Asked of users who: . . . ” followed by a string describing the background of users being asked to answer the question (e.g., “go to Disneyland more than twice a year” or “often answer questions about amusement park rides,” etc.). Other aspects of the inventive subject matter are set forth below in connection with the description of the various figures.
The concept of “following” may be thought of as a unilateral relationship, because it can be achieved at the desire of only one user. This is in contrast to the concept of two users being “connected” (e.g., in a social networking context), which typically implies a personal relationship between the users. Two users may be “connected” when each user is following the other—a bilaterally defined relationship. In some embodiments of the invention, the interactions that the users are allowed to have with one another may be controlled to some extent by the type of relationship that exists between the users, as that relationship is defined by the question-and-answer service, or some externally-hosted social network service. For instance, with some embodiments, a requesting user may only solicit an answer to a question from another user if the requesting user is following the other user, or if the requesting user is being followed by the second user. Similarly, with some embodiments, the requesting user may only be allowed to solicit an answer to a question from another user if the users are connected—that is, each user is following the other. In yet other embodiments, users may be able to solicit an answer from one another regardless of any pre-existing system-defined relationship between the users.
In some examples, a user 10 can also be associated with an interest group. An interest group is a group of users who are interested in discussing a particular subject. The subject can be a topic, such as “amusement parks” or a question, such as “What are the best amusement parks in California?” In some examples, the question-and-answer service can provide a space in which the interest groups can discuss a particular interest. For example, the question-and-answer service can include a message board or forum for the discussion. In some examples, users can start interest groups. In some examples, the question-and-answer service can suggest to the user 10 interest groups that may be of interest to the user, based on the user's activity in the question-and-answer service. In these and other examples, the user 10 can join the interest group, and once the user 10 is a member of the interest group, the user 10 can join the discussion. In some examples, the interest group may be moderated, such that another user (referred to as a moderator) approves each new member of the group. In some examples, members of an interest group can be automatically notified when a user posts a message to the group. In some examples, the user 10 can follow an interest group, instead of or in addition to joining the group. Following the interest group without joining the group can mean, for example, that the user 10 can read messages posted to the group but cannot participate in the discussion.
In some examples, interest groups can also follow users, questions, and/or topics. In these examples, questions and/or answers posted by a user that is followed by the group can automatically be posted to the group discussion, or can be available for viewing from interest group's space. Similarly, answers to the questions and information posted to the topics can be automatically posted to the group discussion and/or be quickly accessible from the group's space.
In some embodiments of the invention, notifications may be communicated to a user by simply including the relevant content in a landing page or data feed displayed to the user. For example, when a user first provides his or her authentication information (e.g., username/password), the user may be presented with a personalized home page or landing page with content that is customized for the user. This content may be selected to include information regarding the users, topics and questions that the particular user is following. Additionally, in some embodiments, notifications may be communicated in near real-time to a user via any number of messaging mechanisms, to include real-time dynamic data feed updates (e.g., on a web page), email, SMS or text messages, instant messages, and others. In some embodiments, a user may elect to be notified of certain activities or events on a per-question, per-topic or per-user basis. For instance, a user may elect to receive a notification when a new answer is posted to a particular question that the user is following.
Referring again to
In various examples, an interest group can also have a summary. The summary can include, for example, a description of the subject that the group is discussing and descriptions of any conclusions or points of contention that the members of the group may have reached. In some examples, the summary can also include information about the members of the group, such as a list of the members, biographical information about the members, and/or an indication of how active each member is in the group discussion. The summary can further server as a repository for any other information about the interest group.
In addition to a summary 18, a question 16 is associated with one or more answers 20. For instance, after a user posts a question, other users of the application are able to provide answers to the question. In some embodiments of the invention, any user is allowed to post an answer to a particular question. As such, a question may be associated with or have multiple answers. In some embodiments, both questions 16 and answers 20 may have comments. For instance, a user may provide a textual comment that is associated with a question 16 or an answer 20. A comment associated with an answer, for example, could provide some clarification about a particular answer, or some aspect of the answer. Other users can then view the comments when viewing the question and/or answers.
In some examples, interest groups can also post answers to a question. For example, the members of the interest group can, within the space provided to them to do so, discuss the question, and determine one answer that can be provided as the answer for the whole group. The interest group's space can include, for example, a voting system that allows the members of the group to vote for an answer that best represents the group's collective knowledge.
In some embodiments of the invention, an answer 20, including one provided by an interest group, has or is associated with votes 24. For example, users can vote up or vote down a particular answer based on whether the user finds the answer helpful in view of the particular question. For instance, if a user believes that a particular answer to a question is a good answer, the user can select a button or other graphical user interface element to vote for the answer. Similarly, if a user believes that a particular answer is not helpful in light of the question, the user can vote down the answer, for example, by simply selecting a button or other graphical user interface element to indicate that the answer is not helpful. As described in greater detail below, with some embodiments, when a requesting user solicits an answer to a question from a second user and/or a group of users, and the second user, a user from the group, or the group as a whole, provides an answer, the requesting user's solicitation may be automatically converted to a vote for the answer. In some embodiments, the number of votes for and against an answer are used as inputs to an algorithm that determines how answers are to be displayed when presented to a user. For example, the votes for and against an answer may simply be tallied, such that a vote for the answer offsets a vote against the answer, and the answers with the highest vote tallies are displayed in the most prominent positions—typically, at the top of a list of relevant answers. With some embodiments, a vote by a particular user may be deemed as more important, for example, as based on the particular user's profile information, and as such, the particular user's vote may be given more weight than a vote by other users.
By providing a forum that includes questions, answers, comments and votes, the question-and-answer application encourages meaningful discussion about a whole host of subject matters, in part, by enabling users to interact with the application in a variety of ways. For instance, some users may desire an entirely passive experience, and can therefore simply browse for, and read, questions and answers on topics of interest. Some users may desire an experience including a moderate level of participation, and as such, these users can vote up or down various answers on topics of interest, and possibly provide commentary. Others may desire to participate more actively, and will elect to post questions and answers to questions—particularly when specifically called upon to do so by another user.
As illustrated in
The question-and-answer application logic 30 is shown in
In some embodiments, some of the various functional components of the question-and-answer application, including some of the various software modules, may be distributed across several server computers, providing application reliability and scalability. For instance, as illustrated in
Consistent with some embodiments, the question-and-answer service is a stand-alone service accessible via its own unique address (e.g., URL or URI). With some embodiments, the stand-alone service may leverage its own social layer, or a social layer provided by an externally-hosted social network service. Accordingly, various relationships between users, as determined or defined by the question-and-answer service or an externally-hosted social network service, may be utilized to customize the functionality and features of the question-and-answer service. For example, certain interactions between users may be controlled based on the relationships that exist between the users, as those relationships are defined by an externally-hosted social network service. Alternatively, with some embodiments, the question-and-answer service may be one of several applications or services that are associated with, and provided by, a social network service. For instance, the question-and-answer service may be accessible via the same address or domain by which users access a social network service, such that the question-and-answer service is hosted by the same entity providing the social network service.
Referring again to
In some embodiments, the user may choose a user or group of users to whom the question is to be directed by searching for a name (e.g., Jacques Cousteau), or for a piece of credential text (e.g., “marine biologist”). In some embodiments, the user may choose a user or group of users to whom the question is to be directed by choosing from a list of algorithmically recommended users. In some embodiments, the user may choose a user or group of users to whom the question is to be directed by selecting a topic that the question is tagged with (e.g., “marine biology”), then choosing from a list of algorithmically recommended users who have written about that topic before or are likely to write about that topic.
Algorithmically recommended users may be presented to the user in the form of a ranked list of users that may be generated by a machine learning algorithm. In some embodiments, the algorithm may be a supervised learning model trained to predict the expected quality score of an answer that the recommended user will write given that the recommended user is shown to the requesting user as a suggestion. In some embodiments, the algorithm may rank recommended users based on a combination of how likely the requesting user is to pick the recommended user, how likely the recommended user is to write an answer in response, and how high quality that answer will be.
In some embodiments, the machine learning algorithm may be a gradient boosted decision tree. This model may take many inputs or features to make good predictions. Exemplary features include whether the recommended user follows the requesting user, the recommended user's past answer count in the topic(s) of the question, the textual similarity between the question at hand and other questions that the recommended user has answered, and the recommended user's historical answer rate in the topic(s) of the question. In some embodiments, not every recommended user is scored by this model, and instead only a set of candidate recommended users may be scored. Candidates may be selected in a variety of ways, such as users who have written answers before in the topic(s) of the question, users who have written about similar topic(s), users who have expressed interest in the topic(s) in other ways (credibility text, past activity, etc.), and users who have answered similar questions.
The ranked list of recommended users may be generated in the same way in some embodiments, except the candidate users may be restricted to users who have written about the chosen topic. In some embodiments, some candidates may also be included who are likely to write about the chosen topic based on their activity (credibility text, past interaction with topics, etc.).
After receiving the user input or user selection identifying the user or group of users to whom the question is to be directed (e.g., USER B in
However, if the user to whom the question is being directed has not previously indicated a desire to pass on answering the question, at method operation 60 the user to whom the question is being directed (e.g., USER B) is sent a message that specifically indicates that the requesting user (e.g., USER A) would like the user (e.g., USER B) to provide an answer to the particular question. The message may be, for example, an e-mail, a mobile push notification, an application notification, and/or the like. The content of the message and the means by which it is communicated may vary depending upon the particular implementation. For instance, with some embodiments, the message may be communicated via a proprietary notification mechanism, such that a notification message is shown on a web page displayed to the user (e.g., USER B). Two examples of such messages are shown, and further described in connection with,
In some embodiments, a notification is not sent to the user to whom the question is being directed. For example, sending a notification for every single answer request could annoy receivers and cause them to stop responding to requests or even to stop using the application. Thus, algorithms may be used to decide which requests a receiver is most likely to be interested in answering and only send notification for those requests. In some embodiments, this may be a machine learning model trained using supervised learning.
In some embodiments, a model may be used that is trained to predict whether a receiver will write an answer if he or she receives a notification. The model may output a score and a notification may be sent if the score is above a cutoff score. This cutoff score may be determined based on the user's past answering behavior. To prevent extreme receiver behavior, each receiver may also have a maximum number of requests that they may receive per week. This “capacity” may also dependent on past answering behavior, i.e., the more they have answered, the higher their capacity.
In some embodiments, a model may be used that is trained to predict whether a receiver will click on a notification. A notification may be sent if the model's output score is above a cutoff score. This cutoff score may be chosen based on data analysis to try to maximize the user's lifetime answers. In some embodiments, reinforcement learning may be used to choose the cutoff score automatically.
These models may be gradient boosted decision tree models in some embodiments. Other than the reinforcement learning cutoff score algorithm, cutoff score and capacity/throttling algorithms may be heuristic algorithms that have been tuned via A/B testing.
Returning to method operation 56, when the question is directed to a group of users, in some examples, when the group has previously indicated a desire to pass on answering the question, then, then, at method operation 58, the user who is soliciting the answer receives a message so indicating. The group of users can be an interest group, in which case the group, as a whole, may have indicated a desire to pass on answering. The group of users can alternatively be users with a similar background, in which case the method proceeds to operation 58 only when all the users in the group have previously passed on answering the question.
With some embodiments, a user can direct a question to any user of the question-and-answer service, and as such, a list of users will be automatically populated with names selected from the entire community of users, so long as those names match the user input. In some alternative embodiments, a user may be limited in directing a question only to some subset of the entire user community, such as those users with whom the user has established a unilateral relationship (e.g., by “following” another user), or a bilateral relationship (e.g., where each user follows the other). In such a scenario, the list of names dynamically updated for selection would include only those names satisfying the system-defined criteria. In some embodiments, a user can enter contact information for a person who is not currently a user of the question-and-answer service. For example, by entering an e-mail address, telephone number, or some other contact information, an appropriate message may be communicated to the person with a request to provide an answer to a question. The recipient of a solicitation message may be prompted to provide his or her name to be displayed with his or her answer, or alternatively, the recipient may be prompted to register as a user of the question-and-answer service as part of the process of providing the answer.
In some examples, the user can, alternatively or additionally, direct a question to any interest group. A list of interest groups can be automatically populated with the names of interest groups. The interest groups in the list can be determined from, for example, interest groups the user is a member of and/or is following and/or interest groups that are following and/or discussing the question and/or topic the question is related to. As another example, an interest group can be added to the list when a number of members of the group (e.g., 75% or another threshold value) have provided answers to the question.
In some examples, the user can, alternatively of additionally, direct the question to a group of users that have a common background. This group of users may not be members of an interest group, and may instead have other commonalities. For example, the users may frequently answer the same or similar questions, follow the similar questions and/or topics, follow the similar users, be members of the same or similar interest groups, and/or have similar information in their user profiles. In various examples, the commonality that is extracted is related to the question and/or topic. In these and other examples, the question-and-answer service can provide a brief description of a group, and display a list of these brief descriptions in the question page. For example, for the question “What are the most popular rides at Disneyland,” the question-and-answer service can identify and display groups of users such as “Users who go to Disneyland more than twice a year,” “Users who often post about Disneyland,” and “Users who used to work at Disneyland,” among others. The user can then select from among the brief descriptions.
In various examples, the question-and-answer service can enable the user to direct the question to multiple individual users and/or multiple groups of users.
At method operation 62, subsequent to communicating a request to answer a question to a user (e.g., USER B), the total number of requests to answer the question that have been received by the user is computed. For instance, the sum of all ask-to-answer requests that have been directed to a particular user for a particular question is computed so that it can be displayed in a notification on the question page for the question. An example of such a notification is shown in
The method begins at method operation 70 when a user (in this example, USER C) requests a question page for a question for which at least one user has already been solicited for an answer from another user (e.g., USER B). For instance, continuing with the example shown in
Consistent with some embodiments, each individual solicitation for an answer to a question is converted to a vote for the answer that is provided by the person who has been solicited for the answer. For instance, consider again the example presented in
In various examples, a method similar to the method of
In examples where the question is directed to an interest group, the interest group can receive a notification of the request to answer the question. For example, the notification can be posted to the group discussion, possibly with the text of the question.
In examples where the question is directed to a group of users with a common background, each of the users in the group can receive a notification in a similar manner as is described above in method operation 80.
When the question is directed to a particular user, then at method operation 82, the question-and answer service receives text representing an answer to the question from the solicited user (e.g., USER B). The text may be received, for example, through the submission of a web-based form displayed to the user via an e-mail or web page. Finally, in some embodiments, at method operation 84, each solicitation for an answer to the question received by the user (e.g., USER B) is converted to a vote for the answer provided by the user. This final method operation 84 is an optional operation and may not be performed in all embodiments.
When the question is directed to a group of users, including interest groups, answers can come from one or more individual users in the group. When the question is directed to an interest group, one answer may be provided by the interest group as a whole.
The method described above is similar when the question is directed to an interest group, and the interest group has determined not to answer the question. For example, the interest group can be presented with a message indicating a solicitation to provide an answer to a question. In some examples, the members of the interest group may reach a consensus not to answer the question, and may so indicate (e.g., a moderator may respond for the group). In some examples, a group moderator or administrator may have authority from the group to decide whether the interest group should pass on answering the question. In these and other examples, when the interest group elects to pass on answering the question, the method can include communicating a message to each user that solicited the interest group to answer the question.
As illustrated in
In this example user interface, beneath the question 104 are two answers 112 and 114 provided by two different users. Associated with each answer is a name of the user who posted the answer, along with an indication of the number of people following that user. For example, the first answer 112 was posted by the user, John Smith, followed by one-hundred-twelve (112) users, while the second answer 114 was posted by Kate Johansen with no followers. Beneath each respective answer are buttons or links that enable a user to add a comment about an answer, or indicate that an answer was helpful or not helpful. In addition, to the left of each answer are two arrows for use in voting. For example, the arrow buttons (up and down) for the answer posted by John Smith are shown with reference number 116. The votes for and against an answer may be used in an algorithm that determines the order in which answers to a particular question are presented.
At the bottom of the example user interface in
In the right side of the user interface, there is a “FOLLOW QUESTION” button 122 that enables a user to subscribe to, or follow, the question. By following the question, the user will be notified when others post new answers to the question. Beneath the “FOLLOW QUESTION” button 122 are several questions determined to be related to the question 104 on the current question page. Selecting one of these questions will lead the user to the question page for the selected question.
In addition, to the right side of the example user interface shown in
Next to the “SHARE QUESTION” button 124 is an “ASK TO ANSWER” button 126. The “ASK TO ANSWER” button 126 enables a user to solicit another user for an answer to the question. For instance, as described in connection with the method illustrated in
Referring now to
In some embodiments, the question-and-answer service will automatically select the list of user's names to be included in the drop-down list 132 based on the user-provided input, and will then rank or order the names based on a variety of factors, such that the names of those users who are most likely to provide a meaningful answer to the question are shown first (e.g., at the top) in the list. For example, the names in the list 132 may be ranked, and then ordered, based on whether a user has a relationship with the requesting or soliciting user, such that those users with whom the requesting user has a relationship are shown in a more prominent position in the list. In some embodiments, those users who have previously answered questions on a topic to which the question being shared is assigned will be shown more prominently in the list. In some embodiments, users who have been recognized as experts on a topic, either through user endorsements or by some other means of recognition, are given greater weight in the ranking algorithm, and are therefore shown more prominently in the list. In some embodiments, users who are currently online—for instance, participating in an active user session with the question-and-answer service—are given greater weight in the ranking algorithm and are therefore shown more prominently in the list. These, and a variety of other factors, may be utilized when presenting a list of names for selection by the soliciting user.
Once a user identifies the name of a user to whom a question is to be directed, the user simply selects the name from the list and the question-and-answer service automatically communicates a message to the selected user, such that the message requests that the selected user provide an answer to the question. In addition, the question-and-answer service stores information associated with the solicitation so that such information can be displayed on the question page for the question in response to subsequent requests for the question page. This information may include, for example, the name of the user that has requested an answer to the question, the name of the user who has been requested to answer the question, the number of users who have requested that a particular user answer the question, and so forth.
In some embodiments, the question-and-service may analyze a variety of factors to recommend or suggest to a particular user the names of other users who may be well-suited to answer a particular question, or highly likely to provide an answer, if asked to provide an answer. For instance, such factors may include any one or more of the following, in various combinations: the relationship that exists between the users, the number of questions on a particular topic that a user has previously answered, the net number of positive votes that user's answers have received for a certain topic, a high correlation between the topics assigned to a question and the topics being followed by a user, and the designation or recognition of a user as a knowledge expert on a given topic. With some embodiments, the suggested users from whom a particular user might solicit an answer are displayed by default on the question page. Alternatively, the question page may include a user interface element (e.g., button or link), which, when selected, generates an additional user interface that presents the suggested users from whom the particular user might solicit an answer to a question. With some embodiments, when displaying the list of suggested users who might be solicited for an answer to a question, a topic-specific biographical statement pertaining to each suggested user may be presented along with the suggested user's name, to provide the viewing user with some additional information for use in assessing whether or not to direct a question to the user.
In some examples, the text input box 130 cans also be used to enter the name of an interest group. In these examples, a list of interest groups can be generated, automatically and in real time, and be displayed in response to text being input into the text input box. In some examples, a separate user interface element can be used for entering the names of interest groups and displaying a list of interest groups.
In some examples, the text input box 130 or another user interface element can be used to enter a description of a group of user to which to direct the question. The user interface element can enable input of text such as, “Parents with kids under 12 who have been to Disneyland” or something similar.
Although not illustrated in
In some examples, the user interface illustrated in
Skilled artisans will recognize that the methods and user interfaces presented herein are presented as specific examples of how the inventive concepts may be implemented. However, methods and user interfaces that differ from the specific examples presented herein may be consistent with the general scope and spirit of the inventive concepts. Particularly, the order in which certain method operations occur may vary from the examples presented herein without departing from the scope and spirit of the inventive concepts. Similarly, a wide variety of user interface elements different from those presented herein may be used to implement the inventive concepts described herein.
The various operations of example methods described herein may be performed, at least partially, by one or more processors that are temporarily configured (e.g., by software) or permanently configured to perform the relevant operations. Whether temporarily or permanently configured, such processors may constitute processor-implemented modules that operate to perform one or more operations or functions. The modules or logical components referred to herein may, in some example embodiments, comprise processor-implemented modules or logic.
Similarly, the methods described herein may be at least partially processor-implemented. For example, at least some of the operations of a method may be performed by one or more processors or processor-implemented modules. The performance of certain of the operations may be distributed among the one or more processors, not only residing within a single machine, but deployed across a number of machines. In some example embodiments, the processor or processors may be located in a single location (e.g., within a home environment, an office environment or as a server farm), while in other embodiments the processors may be distributed across a number of locations.
The one or more processors may also operate to support performance of the relevant operations in a “cloud computing” environment or as a “software as a service” (SaaS). For example, at least some of the operations may be performed by a group of computers (as examples of machines including processors), these operations being accessible via a network (e.g., the Internet) and via one or more appropriate interfaces (e.g., Application Program Interfaces (APIs).)
The example computer system 1500 includes a processor 1502 (e.g., a central processing unit (CPU), a graphics processing unit (GPU) or both), a main memory 1501 and a static memory 1506, which communicate with each other via a bus 1508. The computer system 1500 may further include a display unit 1510, an alphanumeric input device 1517 (e.g., a keyboard), and a user interface (UI) navigation device 1511 (e.g., a mouse). In one embodiment, the display, input device and cursor control device are a touch screen display. The computer system 1500 may additionally include a storage device (e.g., drive unit 1516), a signal generation device 1518 (e.g., a speaker), a network interface device 1520, and one or more sensors 1521, such as a global positioning system sensor, compass, accelerometer, or other sensor.
The drive unit 1516 includes a machine-readable medium 1522 on which is stored one or more sets of instructions and data structures (e.g., software 1523) embodying or utilized by any one or more of the methodologies or functions described herein. The software 1523 may also reside, completely or at least partially, within the main memory 1501 and/or within the processor 1502 during execution thereof by the computer system 1500, the main memory 1501 and the processor 1502 also constituting machine-readable media.
While the machine-readable medium 1522 is illustrated in an example embodiment to be a single medium, the term “machine-readable medium” may include a single medium or multiple media (e.g., a centralized or distributed database, and/or associated caches and servers) that store the one or more instructions. The term “machine-readable medium” shall also be taken to include any tangible medium that is capable of storing, encoding or carrying instructions for execution by the machine and that cause the machine to perform any one or more of the methodologies of the present invention, or that is capable of storing, encoding or carrying data structures utilized by or associated with such instructions. The term “machine-readable medium” shall accordingly be taken to include, but not be limited to, solid-state memories, and optical and magnetic media. Specific examples of machine-readable media include non-volatile memory, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks such as internal hard disks and removable disks; magneto-optical disks; and CD-ROM and DVD-ROM disks.
The software 1523 may further be transmitted or received over a communications network 1526 using a transmission medium via the network interface device 1520 utilizing any one of a number of well-known transfer protocols (e.g., HTTP). Examples of communication networks include a local area network (“LAN”), a wide area network (“WAN”), the Internet, mobile telephone networks, Plain Old Telephone (POTS) networks, and wireless data networks (e.g., Wi-Fi™ and WiMax™ networks). The term “transmission medium” shall be taken to include any intangible medium that is capable of storing, encoding or carrying instructions for execution by the machine, and includes digital or analog communications signals or other intangible medium to facilitate communication of such software.
Similarly,
Although an embodiment has been described with reference to specific example embodiments, it will be evident that various modifications and changes may be made to these embodiments without departing from the broader spirit and scope of the invention. Accordingly, the specification and drawings are to be regarded in an illustrative rather than a restrictive sense. The accompanying drawings that form a part hereof, show by way of illustration, and not of limitation, specific embodiments in which the subject matter may be practiced. The embodiments illustrated are described in sufficient detail to enable those skilled in the art to practice the teachings disclosed herein. Other embodiments may be utilized and derived therefrom, such that structural and logical substitutions and changes may be made without departing from the scope of this disclosure. This Detailed Description, therefore, is not to be taken in a limiting sense, and the scope of various embodiments is defined only by the appended claims, along with the full range of equivalents to which such claims are entitled.
Cole, David, Sharma, Abhinav, D'Angelo, Adam Edward, Mazhari, Muhammad Emmad
Patent | Priority | Assignee | Title |
Patent | Priority | Assignee | Title |
6064978, | Jun 24 1997 | EXPERTS EXCHANGE LLC | Question and answer system using computer networks |
6433795, | Nov 08 1996 | Meta Platforms, Inc | System for integrating an on-line service community with a foreign service |
7523167, | Jan 16 2004 | TRAVELCLICK, INC | Systems and methods for optimizing dynamic mailings |
7957510, | Jan 23 2007 | Microsoft Technology Licensing, LLC | Notifying network contacts of inquiries |
8516379, | Dec 31 2010 | QUORA, INC | Methods and systems for soliciting an answer to a question |
8751559, | Sep 16 2008 | Microsoft Technology Licensing, LLC | Balanced routing of questions to experts |
9124447, | Jul 26 2002 | International Business Machines Corporation | Interactive client computer communication |
9183251, | Jan 09 2012 | GOOGLE LLC | Showing prominent users for information retrieval requests |
20010032244, | |||
20030140037, | |||
20050125544, | |||
20060240851, | |||
20070219958, | |||
20090012833, | |||
20090049042, | |||
20090125598, | |||
20100070554, | |||
20110041082, | |||
20110055207, | |||
20110055309, | |||
20110196922, | |||
20110246910, | |||
20110302098, | |||
20120173992, | |||
20130339879, | |||
20140032273, | |||
20150019525, | |||
20160259789, | |||
20200089708, |
Executed on | Assignor | Assignee | Conveyance | Frame | Reel | Doc |
May 06 2019 | Quora, Inc. | (assignment on the face of the patent) | / | |||
May 30 2019 | D ANGELO, ADAM EDWARD | QUORA, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 049398 | /0695 | |
May 30 2019 | SHARMA, ABHINAV | QUORA, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 049398 | /0695 | |
May 30 2019 | MAZHARI, MUHAMMAD EMMAD | QUORA, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 049398 | /0695 | |
May 30 2019 | COLE, DAVID | QUORA, INC | ASSIGNMENT OF ASSIGNORS INTEREST SEE DOCUMENT FOR DETAILS | 049398 | /0695 |
Date | Maintenance Fee Events |
May 06 2019 | BIG: Entity status set to Undiscounted (note the period is included in the code). |
May 16 2019 | SMAL: Entity status set to Small. |
Date | Maintenance Schedule |
Feb 08 2025 | 4 years fee payment window open |
Aug 08 2025 | 6 months grace period start (w surcharge) |
Feb 08 2026 | patent expiry (for year 4) |
Feb 08 2028 | 2 years to revive unintentionally abandoned end. (for year 4) |
Feb 08 2029 | 8 years fee payment window open |
Aug 08 2029 | 6 months grace period start (w surcharge) |
Feb 08 2030 | patent expiry (for year 8) |
Feb 08 2032 | 2 years to revive unintentionally abandoned end. (for year 8) |
Feb 08 2033 | 12 years fee payment window open |
Aug 08 2033 | 6 months grace period start (w surcharge) |
Feb 08 2034 | patent expiry (for year 12) |
Feb 08 2036 | 2 years to revive unintentionally abandoned end. (for year 12) |